Checked output
UTF-8 or GB18030 input, consistent column counts, preserved text identifiers, and a separate output file.
Fixed scope · proposed $25
Clean one CSV with explicit, reviewable rules.
Receive a separate cleaned file, reproducible Python script, and before/after report. Leading-zero identifiers, quoted commas, Chinese text, and the original source are deliberately protected.
Use fictional examples only in a public issue. A quote request is not an order, and no payment is requested before the scope is agreed.
Agreed duplicate rule
The default demonstration removes only exact duplicate rows. Any business-key rule is written and approved first.
Reproducible delivery
The cleaned CSV is accompanied by the script used and a JSON report describing counts and changes.
Proposed $25 scope
- One CSV, up to 20,000 rows and 20 columns
- UTF-8 or GB18030 input
- Surrounding-whitespace cleanup
- One explicitly agreed duplicate rule
- One correction within the agreed scope
Confirmed before ordering
Exact transformations, acceptance criteria, deadline, private file exchange, delivery method, and payment instructions are confirmed first. XLSX conversion, inferred business rules, research, enrichment, and additional datasets require a separate proposal.
Privacy boundary: do not upload customer records, personal data, credentials, or payment details to GitHub. The public issue should contain only dimensions, rule descriptions, and fictional examples. Private data handling must be agreed separately.
Questions before requesting a quote
What does the $25 proposal include?
One file within the stated size limits, the agreed cleanup, cleaned output, reproducible script, change report, and one correction within the agreed scope.
Can I post my real CSV in the issue?
No. Use fictional examples only. Private data exchange and handling must be agreed before the order is accepted.
When do I pay?
Only after scope, acceptance criteria, timing, private exchange, payment instructions, and delivery arrangements are agreed.